© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 30544
    Prev
    Next
    INDEX
    Explanations

    conjunctions and coordinate multiple phrases or clauses

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    191
    +0.14
    0.6%
    199
    +0.09
    0.4%
    286
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.14
    0.11
    746
    +0.09
    0.04
    496
    +0.09
    0.09
    Negative Logits
    cised
    -0.71
    ledged
    -0.69
    scribe
    -0.68
    prus
    -0.64
    $$$$
    -0.63
    aceae
    -0.63
    ukong
    -0.60
    krit
    -0.60
    ?????-
    -0.60
    tie
    -0.60
    POSITIVE LOGITS
     distributing
    1.61
     gaining
    1.59
     maintaining
    1.58
     making
    1.58
     creating
    1.57
     preventing
    1.57
     delivering
    1.57
     discovering
    1.57
     improving
    1.56
     avoiding
    1.54
    Activations Density 1.174%

    No Known Activations